Perhaps 300ms reaction time is too inhuman, but looks like BBF's prediction was spot on.

Also, what's worth noting is BBF in his 42:51 performs this skip in about 13.3s, which is only a little over a second slower to this already ridiculous human theory estimate. And the way these pause buffers work, going for a 1 frame might result in 0 or 2 frames.

It takes Anso 24s to do the same skip in the world record - without retries. To be able to do it in 13s, which is on par with doing it perfectly using the save - in a run no less - is difficult to wrap my mind around.

New comparison vs the new strat. Both human execution this time - been trying a different way to pause buffer with decent results. Obviously pretty lucky, so still rather severely unviable for TA, but it's interesting how fast you can realistically go with this. Took Anso 20s in his 39:28 with uncomfortable keybinds.